---
title: "Portfolio"
output: 
  flexdashboard::flex_dashboard:
    theme:
      bg: "#101010"
      fg: "#FDF7F7" 
      primary: "#194d33"
      base_font:
        google: Prompt
      code_font:
        google: JetBrains Mono
    orientation: columns
    vertical_layout: fill
    source_code: embed
---

```{r setup, include=FALSE, cache=TRUE}
library(flexdashboard)
require(mosaic)
require(ggplot2)
require(readr)
require(tidyr)
require(socviz)
require(maps)
require(dplyr)
require(broom)
require(raster)
require(viridis)
require(Lock5Data)
require(forcats)
theme_set(theme_bw())
# Install thematic and un-comment for themed static plots (i.e., ggplot2)
# thematic::thematic_rmd()

## DATA IMPORTS

#Graph Refinements
collegeDistance <- read_csv("~/CSVs/CollegeDistance.csv")

#Maps
data("election")
election$region = tolower(election$state)

States = map_data("state")

FullData = left_join(States,election)

all_county_data = left_join(county_map,county_data)

#Time
data("Births2015")


#Dashboard
data("BaseballHits2019")
BaseballHits2019$Wins <- as.numeric(BaseballHits2019$Wins)
BaseballHits2019$Team <- as.factor(BaseballHits2019$Team)
AL = subset(BaseballHits2019, League == "AL")
NL = subset(BaseballHits2019,League == "NL")
team_wins = gf_point(Wins~fct_reorder(Team,Wins),data=AL,color=~Team) %>%
  gf_theme(axis.text.x = element_text(angle = 45, hjust = 1),legend.position="none") %>%
  gf_labs(x="Teams")
```
Graph Refinements
=======================================================================


Column {data-width=500}
-----------------------------------------------------------------------
### Original Graph
```{r, cache=TRUE}
gf_boxplot(score~gender,data=collegeDistance, fill=~gender)
```


Column {data-width=500}
-----------------------------------------------------------------------
### Refined Graph
```{r, cache=TRUE}

means = data.frame(collegeDistance %>%
  group_by(gender, ethnicity) %>%
  summarise(mean_score = mean(score)))

gf_boxplot(score ~ gender,data = collegeDistance, fill =  ~ gender,alpha = 0.75) %>% 
  gf_facet_wrap( ~ ethnicity, nrow = 3, ncol = 2) %>%
  gf_labs(title="Gender and Ethnicity Correlate to Different Average Test Scores",
    x="Gender",y="Scores") +
  geom_violin(alpha = 0.4, color = "grey30") +
  geom_text(data = means, aes(x = gender, y = mean_score, label = sprintf("Mean:\n%.2f", 
    mean_score), color = gender),
    position = position_dodge(width = 0.8), vjust=-3, hjust=2, size=3)
```


Graphs with Time
=======================================================================


Column {data-width=500}
-----------------------------------------------------------------------
### Montly Tiles
```{r, cache=TRUE}
# Added Shift Variable
Births2015.shift = subset(Births2015,day_of_month==1) %>%
  mutate(Shift = day_of_week - day_of_month) %>%
  dplyr::select(month,Shift)

# Joined into new dataset
new.Births2015 = full_join(Births2015, Births2015.shift) %>%
  mutate(week_of_month = ceiling((day_of_month + Shift)/7))


gf_tile(week_of_month ~ wday | factor(month.abb[month],levels=month.abb), data=new.Births2015,
        fill = ~births, show.legend = FALSE) %>%
  gf_refine(scale_y_reverse(),scale_fill_viridis(), 
            scale_x_discrete(labels = function(x) abbreviate (x, minlength=1))) %>%
  gf_theme(axis.text.y= element_blank(), axis.ticks.y = element_blank(),theme_light()) %>%
  gf_labs(x="",y="",title="Most Births Happen on Weekdays", subtitle= "With the exception of US Holidays")
```

Column {data-width=500}
-----------------------------------------------------------------------
### Before
```{r, cache=TRUE}
gf_line(births~date, data=Births2015) %>%
  gf_refine(coord_cartesian(xlim=c(as.Date("2015-01-01"), as.Date("2015-4-30")))) %>%
  gf_refine(scale_x_date(date_minor_breaks = "1 week")) %>% gf_labs(title="Most Births Happen in the Beginning of the Week.", x="Time", y="Number of Births")
```

### After
```{r, cache=TRUE}
gf_line(births~date, data=Births2015, color=~wday) %>% 
  gf_labs(title="Most Births Happen During the Week compared to Weekends.", x="Time",
    y="Number of Births",color="Day of the Week")
```


Mapping Data
=======================================================================

Column {data-width=550}
-----------------------------------------------------------------------

### Israel Earthquakes Map

```{r, cache=TRUE}
israel = getData('GADM', country='ISR', level=1)
israel_map = tidy(israel)
Earthquake = read_csv("~/CSVs/Earthquake.csv")
Earthquake = Earthquake %>%
  rename(LocationName= `Location Name`)
Earthquake$LocationName = tolower(Earthquake$LocationName)
 gf_point(Latitude~Longitude,data=subset(Earthquake, grepl("israel",LocationName)), size=1,color="red") %>%
gf_polygon(lat~long,data=israel_map,group=~group,alpha=0.75) %>%
  gf_refine(coord_equal()) %>%
   gf_labs(title="\nMost Earthquakes in the Israel Region\n\tHappen in the West Bank")
```

Column {data-width=450}
-----------------------------------------------------------------------

### US Counties

```{r, cache=TRUE}
gf_polygon(lat~long,data=all_county_data, group=~group,fill=~winner)%>%
  gf_theme(theme_map()) %>% 
  gf_refine(scale_fill_manual(values = c("royalblue","red")), coord_equal()) %>%
  gf_labs(fill="Winner", title="\n\tTrump Wins the 2016 Election")
clinton_county_votes = nrow(subset(all_county_data %>% dplyr::distinct(name, winner), winner != "Trump"))
trump_county_votes = nrow(subset(all_county_data %>% dplyr::distinct(name, winner), winner == "Trump"))
```

### Valuebox1

```{r, cache=TRUE}
valueBox(format(clinton_county_votes,big.mark=","), caption="Clinton County Wins", color="blue",icon="fa-arrow-down")
```

### Valuebox2
```{r, cache=TRUE}
valueBox(format(trump_county_votes,big.mark=","), caption="Trump County Wins",color="red", icon="fa-arrow-up")
```

Baseball 2019 Dashboard
=======================================================================

Column {data-width=400}
-----------------------------------------------------------------------

### Baseball Hits

```{r, cache=TRUE}
BaseballHits2019v2 = gather(BaseballHits2019,key="HitTypes",value="HitValue",6:8)
hits = gf_boxplot(HitValue~HitTypes,data=BaseballHits2019v2,color=~HitTypes) %>%
  gf_theme(axis.text.x = element_text(angle = 45, hjust = 1),legend.position="none") %>%
  gf_labs(x="Hit Types", y="Count")
ggplotly(hits)
```

Column {data-width=400}
-----------------------------------------------------------------------


### Baseball Base Stealing
```{r, cache=TRUE}
stolen = gf_point(StolenBases~CaughtStealing,data=BaseballHits2019,color=~League) %>%
  gf_theme(axis.text.x = element_text(angle = 45, hjust = 1),legend.position="none") %>%
  gf_labs(x="Number of Stolen Bases", y="Number of Times Caught Stealing")
ggplotly(stolen)
```


Column {data-width=175}
-----------------------------------------------------------------------
### Stats
```{r, cache=TRUE}
valueBox(round(mean(BaseballHits2019$Doubles),2),caption="Average # of Doubles",icon = "fa-hourglass-half",color="lightblue")
```

### Stats
```{r, cache=TRUE}
valueBox(round(mean(BaseballHits2019$Triples),2),caption="Average # of Triples",icon = "fa-fast-forward",color="info")
```


### Stats
```{r, cache=TRUE}
valueBox(round(cor(StolenBases~CaughtStealing,data=BaseballHits2019),2),caption="Correlation Between Stealing and Caught Stealing",icon = "fa-line-chart",color = "warning")
```

### Stats
```{r, cache=TRUE}
valueBox(sum(AL$StolenBases),caption="American League # of Stolen Bases",icon = "fa-hand-paper-o",color="danger")
```


### Stats
```{r, cache=TRUE}
valueBox(sum(NL$StolenBases),caption="National League # of stolen bases",icon = "fa-hand-paper-o",color="blue")
```


Hightlights
=======================================================================

Column {data-width=500}
-----------------------------------------------------------------------
### Things I Have Learned and Enjoyed

1. How to represent data in the forms of dashboards.
1. How to wrangle or reorganize data via the gather function.
1. How to create interactive plots via plotly


Column {data-width=500}
-----------------------------------------------------------------------

### Things I Was Confused About

1. Data over time and how to represent in a graph
1. Understanding how Datasets in Handbook of Small Datasets was organized